d-Amino Acid Mediated Recruitment of ... - ACS Publications

Publication Date (Web): May 28, 2014 ... This system represents a novel strategy as an antibacterial therapy that targets planktonic Gram-positive bac...
0 downloads 0 Views 3MB Size
Download PDF
Articles pubs.acs.org/acschemicalbiology

D‑Amino

Acid Mediated Recruitment of Endogenous Antibodies to Bacterial Surfaces Jonathan M. Fura, Mary J. Sabulski, and Marcos M. Pires* Department of Chemistry, Lehigh University, Bethlehem, Pennsylvania 18015, United States S Supporting Information *

ABSTRACT: The number of antibiotic resistant bacterial strains has been continuously increasing over the last few decades. Nontraditional routes to combat bacteria may offer an attractive alternative to the ongoing problem of drug discovery in this field. Herein, we describe the initial framework toward the development of bacterial D-amino acid antibody recruitment therapy (DART). DART represents a promising antibiotic strategy by exploiting the promiscuity of bacteria to incorporate unnatural Damino acids and subsequently recruit antibodies to the bacterial surface. The conjugation of 2,4-dinitrophenyl (DNP) to various D-amino acids led to the discovery of a D-amino acid that specifically tags the surface of Bacillus subtilis and Staphylococcus aureus for the recruitment of anti-DNP antibodies (a highly abundant antibody in human serum). This system represents a novel strategy as an antibacterial therapy that targets planktonic Gram-positive bacteria.

insights into many biological processes such as cell division and bacteria−host interactions.10 A myriad of approaches have been explored to decorate the surface of both Gram-positive and Gram-negative bacteria, including the use of unnatural sugars, 11−13 cell wall components,14−16 and enzymatic modification to the oligopeptides.17−22 Recently, it was revealed that diverse species of bacteria can readily incorporate unusual D-amino acids (e.g., Dmethionine, D-valine, D-tyrorine) into their peptidoglycan from the surrounding medium by replacing the terminal D-alanine of the oligopeptide chain.21 All species of bacteria analyzed to date have been found to possess this feature, an indication that the process of D-amino acid incorporation may be inherent to all peptidoglycan-containing bacteria. The incorporation of the Damino acid has since been attributed to a promiscuous enzymatic activity of bacterial D,D-transpeptidases.23 These enzymes seemingly show little discrimination in terms of side chain selectivity of the D-amino acid, an unusual property for most enzymes.17,19,20 To date, the use of D-amino acids to remodel the surface of bacteria with the goal of therapeutic interventions has not yet been explored. We set out to develop a potentially novel antimicrobial strategy by exploiting the propensity of bacteria to incorporate exogenous D-amino acids into their peptidoglycan that are

For many decades, the number of new antibiotics introduced into clinics easily outpaced the emergence of resistant strains. Infections that could have been life-threatening were quickly cleared with antibiotics. Unfortunately, the current outlook appears much bleaker, with the severe paucity of novel antibiotics discovered in the past 50 years giving the upper hand back to pathogenic bacteria.1−3 Furthermore, the rampant, and sometimes unregulated, use of FDA-approved antibiotics has directly contributed to a rapid rise in the number of antibiotic resistant bacteria.4 With the culmination of the Golden Age of antibiotic treatment, the question as to what novel antibacterial therapies may be used in the future remains to be answered.5,6 The cell wall of Gram-positive bacteria has served as a fruitful target for the development of antibacterial therapies for decades.7 Gram positive bacteria are surrounded by a meshlike scaffold of peptidoglycan, which is composed of multiple layers of N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) glycan strands that are cross-linked by oligopeptides (Figure 1a and b).8 Peptidoglycan provides bacteria with immense structural support. Due to the critical role of the cell wall during cellular growth and division, molecules that interfere with the necessary enzymatic processes related to peptidoglycan synthesis have become some of the most successful antibiotic agents.9 In addition, the chemical remodeling of the surface of bacteria has drawn considerable attention as of late due to its potential to reveal fundamental © 2014 American Chemical Society

Received: January 10, 2014 Accepted: May 28, 2014 Published: May 28, 2014 1480

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

Figure 1. (a) Schematic representation of the peptidoglycan on the surface of a Gram-positive bacterium. (b) Structure of the basic unit of the peptidoglycan. (c) Schematic representation of DART. The bacterial surface is labeled with D-amino acids conjugated to a DNP epitope. Following bacterial cell surface modification, the binding of anti-DNP antibodies leads to the eventual clearance by the immune system.

dinitrophenyl (DNP) epitope into the peptidoglycan (Figure 1c). Although DNP is not naturally produced by humans, approximately one percent of the circulating antibody pool in the human bloodstream binds to DNP epitopes.31−35 We show that when DNP is conjugated to the appropriate D-amino acid backbone, there is both efficient incorporation and subsequent recruitment of anti-DNP antibodies.

tagged with antibody-recruiting antigens. We hypothesized that a synthetic immunology approach could provide a highly efficient method for developing alternative antibiotic agents. The human immune system is adept at combating the vast majority of pathogenic bacteria it encounters. During instances of bacterial infection, bacteria manage to elude the actions of the immune response of the host. Instead of using cytotoxic small molecules to directly destroy bacteria, a strategy that has recently yielded few successful agents, we hypothesized that a selective immuno-modulator could potentially function to reactivate the host immune system to seek out and clear pathogenic bacteria. The field of synthetic immunology via small molecules may be in its infant stages, but it already shows great potential for the development of next-generation immune-modulating therapies. Recent efforts in this field have led to the development of synthetic antigen-displaying molecules to recruit endogenous antibodies to promote the macrophagemediated elimination of HIV particles and have also shown promising results as novel anticancer agents.24−30 Herein, we describe the successful development of the framework for bacterial D-amino acid antibody recruitment therapy (DART). DART relies on the bacterial cell surface remodeling via the incorporation of D-amino acids modified with the 2,4-

■

RESULTS AND DISCUSSION Fluorescent Tracking of D-Amino Acid Incorporation. Initially, we sought to determine the level to which transpeptidases can incorporate D-amino acids into the peptidoglycan of the Gram-positive bacterium B. subtilis (Figure 2a) and the relative selectivity of the stereoisomers by conjugating the small fluorophore 4-chloro-7-nitro-2,1,3-benzoxadiazole (NBD) to D- and L-lysine. The species B. subtilis is widely utilized as a model Gram-positive bacterium due to its ease of handling and its similarity to other pathogenic bacteria.36 The mutant strain B. subtilis ΔdacA, which lacks the D-alanyl-Dalanine carboxypeptidase, has previously been used to yield higher D-amino acid incorporation levels and, as such, was the strain of choice for our experiments.17,20,37,38 Both D-Lys(NBD) and L-Lys(NBD) were synthesized by reacting the NBoc protected amino acids with the amino reactive NBD-F, 1481

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

relative incorporation levels of the amino acid and the stereospecificity of the remodeling strategy using flow cytometry. B. subtilis cells were incubated with either DLys(NBD) or L-Lys(NBD) during the exponential growth phase and analyzed for the acquisition of green fluorescence from the NBD handle (Figure 2b). The flow cytometry data shows that at a concentration of 250 μM of the fluorescent amino acid, the bacteria exposed to D-Lys(NBD) displayed a 30-fold increase in fluorescence signal compared to its enantiomeric L-Lys(NBD) counterpart and a 100-fold increase over unlabeled B. subtilis (Figure S1, Supporting Information). The significant difference between the fluorescence levels in bacteria treated with the enantiomeric molecules suggests that a specific process (in this case, we presume it to be the transpeptidase activity) is selecting out one compound over another. The low levels of fluorescence associated with LLys(NBD) treated bacteria may have resulted from nonspecific binding of the amino acid to the bacterial surface or from low levels of intracellular accumulation. Notably, we observed significant differences in fluorescence levels between the two amino acids at much lower concentrations (10 μM) (Figure S2, Supporting Information). We also observed that the transpeptidases appear to be saturated at a concentration around 1 mM of D-Lys(NBD) as the fluorescence levels start to plateau at this concentration (Figure S2, Supporting Information). The initial discovery that transpeptidases possess the unique ability to accept unusual D-amino acids was confirmed by competition experiments with the natural substrate D-alanine.39 Likewise, we performed competition labeling experiments using D-alanine with bacterial cells incubated with D-Lys(NBD) (Figure S3, Supporting Information). When cross-linking the peptide chains of peptidoglycan, transpeptidases form an acylenzyme intermediate with the donor strand that can either be cross-linked to the acceptor strand or react with a D-amino acid to reform the pentapeptide chain. We observed that the labeling of cells with D-Lys(NBD) was effectively inhibited by increasing the concentration of D-alanine, which is consistent with the incorporation of the D-amino acid via transpeptidases. Next, time-course analysis of the process was conducted to determine the incubation time required for adequate incorporation of the unnatural amino acid. We observed that significant levels of peptidoglycan modification can be detected within 20 min of bacterial incubation with the D-amino acid fluorophore with increasing levels of labeling through the exponential growth phase (Figure 2c). Finally, we also observed efficient incorporation of D-Lys(NBD) by the pathogenic S. aureus at both 250 μM and 1 mM, consistent with previous reports (Figure S4, Supporting Information). Together, these data show that there is rapid and extensive incorporation of DLys(NBD) under a wide-set of conditions. Remodeling of Bacterial Surface with DNP Moieties. Having determined working concentrations and incubation times for D-amino acid integration into the peptidoglycan, we set out to explore the possibility of using amino acids modified with the epitope DNP to recruit antibodies to the surface of the targeted bacteria. DNP was chosen due to the naturally high concentration of anti-DNP antibodies in human serum and its small size.35 In designing a D-amino acid that could recruit endogenous antibodies, we envisioned two distinct factors that would have to be carefully balanced: epitope accessibility and incorporation efficiency. First, in order to be available for antibody binding, the DNP epitope needs to be accessible to the extracellular space. The peptidoglycan of Gram-positive

Figure 2. (a) Schematic representation of the replacement of the terminal D-Ala by the synthetic NBD-displaying D-amino acid DLys(NBD). (b) Flow cytometry analysis of B. subtilis incubated for 4 h in LB alone and LB supplemented with either 250 μM of D-Lys(NBD) or 250 μM of L-Lys(NBD). Inset, DIC and fluorescence microscopy imaging shows the delineation of green fluorescence in B. subtilis labeled with D-Lys(NBD). Scale bar represents 3 μm. (c) Flow cytometry analysis of B. subtilis incubated at various times with 250 μM of D-Lys(NBD). Data are represented as mean ± SD (n = 3).

followed by deprotection with trifluoroacetic acid (Supporting Information). At first, B. subtilis cells were incubated in medium supplemented with D-Lys(NBD) and visualized via fluorescence microscopy (Figure 2b, inset). We observed a uniform fluorescence signal through the entire cell, a finding that is consistent with the remodeling of the cell surface using fluorescent D-amino acids.17 Next, we set out to determine 1482

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

Figure 3. (a) Structures of the DNP displaying D-amino acids synthesized and evaluated for anti-DNP antibody recruitment. (b) Anti-DNP recruitment to the surface of B. subtilis evaluated by flow cytometry. Cells were incubated for 4 h in LB supplemented with 250 μM of specified Damino acids followed by incubation with anti-DNP antibodies. Data are represented as mean + SD (n = 3). Inset, DIC and fluorescence microscopy imaging of cells labeled with 1 mM D-Lys(DNP) followed by anti-DNP. Scale bar represents 3 μm.

of the D-amino acid incorporation by bacteria, we chose to construct a small panel of DNP-displaying D-amino acids. Our D-amino acid panel was designed to encompass molecules of varying linker lengths between the amino acid α-carbon and the DNP epitope (Figure 3a). All compounds were synthesized and purified using standard solution phase or solid phase peptide chemistry (Supporting Information). For this set of experiments, B. subtilis cells were labeled through the exponential phase in medium containing the DNP-displaying Damino acid. Following the initial peptidoglycan remodeling period of 4 h, the bacteria were washed and incubated with Alexa Fluor 488-conjugated anti-DNP antibodies. The ability of each compound to recruit anti-DNP antibodies to the surface of the bacteria was quantified by monitoring the fluorescence levels using flow cytometry (Figure 3b). We observed a distinct pattern of antibody recruitment, whereby amino acids with shorter or longer linker lengths recruit more poorly than the intermediary D-Lys(DNP). While antibody binding was still effectively mediated by the shorter linker D-2,4 diaminobutyric acid (DNP) (D-Dab(DNP)) and D-Ornithine(DNP) (DOrn(DNP)), we observed a loss of antibody recruitment for amino acids with longer linkers. These findings are consistent

bacteria can range in thickness from 20 to 80 nm for mature peptidoglycan.40 The pores within the peptidoglycan have been estimated to be between 5 and 25 nm, the same range as the dimension of an antibody.41 Due to the possibility that antibodies permeate poorly into the peptidoglycan, we anticipated that only surface exposed DNP epitopes would be highly available for antibody recruitment. One possible way to facilitate the presentation of the DNP epitope to antibodies in the extracellular medium is to anchor the DNP onto a long flexible linker connected to a D-amino acid unit. However, it is important to also consider the second and possibly opposing factorthe incorporation efficiency into the peptidoglycan. Although it is evident from past studies that bacteria have some inherent flexibility to the incorporation of various D-amino acids (e.g., large fluorophores), we presumed that there are structural limitations in regards to the size and physical characteristics of the amino acid side chain. The efficiency of incorporation will dictate the total number of D-amino acids found within the peptidoglycan scaffold, which will subsequently affect the capacity of the bacterial surface to recruit anti-DNP antibodies. Due to our limited understanding of the side chain tolerability 1483

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

with our hypothesis that long and flexible side chains reduce incorporation efficiency, but shorter linker lengths reduce the accessibility of the DNP epitope within the peptidoglycan scaffold. D-Lys(DNP) displays the highest antibody recruitment capacity, presumably because it represents a balance of the two competing factors of accessibility and incorporation. Next, we evaluated the antibody-recruitment ability of D-Lys(DNP) at concentrations approaching near the maximum incorporation levels (1 mM). We observed a marked increase in fluorescence levels with the labeling of bacteria at this concentration, consistent with higher levels of D-Lys(DNP) displayed on the bacterial surface (Figure S5, Supporting Information). The fluorescence level observed under these conditions is indicative of robust antibody binding to surface-anchored DNP, especially considering the size of individual B. subtilis cells. Even at 1 mM of amino acid, we observed minimum bacterial growth inhibition (Figure S6, Supporting Information). Finally, we set out to visualize the anti-DNP antibody binding to the surface of DNP-decorated bacteria using fluorescence microscopy (Figure 3b, inset). We observed antibody binding throughout the surface of the cell, but with increased binding at the central portion of the cell compared to either pole. The relative lower levels of surface-bound antibodies at the site of daughter cell separation during cell division are consistent with the location of new peptidoglycan synthesis (Figure S7, Supporting Information).42 The incubation period (1 h) of bacteria with the antibodies at physiological temperatures is long enough to allow partial or complete cell division. This division (or cell elongation) would take place in medium lacking D-Lys(DNP); therefore, the newly synthesized cell wall would not retain the ability to recruit antibodies in our assay conditions. Clearly, the availability of DNP epitopes to the extracellular medium is a critical factor in determining antibody recruitment. We presume that the longer linker length of D-Lys(DNP) led to higher antibody recruitment than in bacteria decorated with the shorter linker D-Dab(DNP). To probe the DNP accessibility further, we first incubated the bacteria in the presence of the DNP-modified D-amino acids at 37 °C and then performed the binding step at two different temperatures, 4 and 37 °C (Figure 4). Basic thermodynamic principles would predict weaker binding between anti-DNP antibodies and surface-bound DNP epitopes at physiological temperatures relative to 4 °C. Yet, we observed that the recruitment of anti-DNP antibodies to bacteria labeled with D-Lys(DNP) was more prominent at the higher temperature. We hypothesize that, at the higher temperature, the polymeric chains within the peptidoglycan can display greater molecular mobility. In turn, either the DNP epitopes on the muropeptide become more accessible to the surrounding medium or the anti-DNP antibodies gain greater permeability into the peptidoglycan scaffold. Alternatively, it may be possible that the anti-DNP antibody used in this study was evolved and functions optimally at 37 °C. Consequently, antibody binding to the DNP epitopes on the cell surface may be more favorable at 37 °C. The ability of D-Lys(DNP)-treated bacteria to bind anti-DNP antibodies and the D-Ala competition experiments are strong indicators that D-Lysine derivatives are likely to be anchored onto the surface by the enzymatic action of transpeptidases. However, these experiments do not definitively show that the D-amino acids are incorporated into the peptidoglycan. In order to confirm that there is covalent modification of the cell wall, we isolated the peptidoglycan from bacteria and digested the

Figure 4. Temperature dependence on anti-DNP recruitment to the surface of B. subtilis cells evaluated by flow cytometry. Cells were incubated for 4 h in LB alone or LB supplemented with 1 mM of DLys(DNP). Following the labeling period, bacteria were incubated with Alexa Fluor 488-conjugated anti-DNP antibodies at 4 and 37 °C and analyzed via flow cytometry. Data are represented as mean + SD (n = 3).

glycosidic linkages using muramidase.43 The crude muropeptides were separated on a RP-HPLC column and fractions displaying an absorbance consistent with the presence of DNP were collected. MALDI-TOF mass spectrometry analysis of the collected sample confirmed the incorporation of the unnatural D-Lys(DNP) and D-Dab(DNP) (Figures S8 and S9, Supporting Information). Next, we utilized the peptidoglycan isolation procedure as a way to quantify relative incorporation levels of the D-amino acid derivatives. We were particularly interested in analyzing the relative incorporation of D-Dab(DNP), DLys(DNP), and D-Lys-PEG1(DNP) due to the fact that we observed different antibody-recruitment among these three amino acids. We hypothesized that the increased accessibility afforded by the longer linker of D-Lys(DNP) relative to DDab(DNP) led to higher recruitment levels, whereas the poorer incorporation of D-Lys-PEG1(DNP) was responsible for decreased antibody recruitment. Using the DNP absorbance as a handle, we found that D-Dab(DNP) and D-Lys(DNP) displayed comparable incorporation levels, while incorporation levels were much lower for D-Lys-PEG1(DNP) (Figure 5). Together, these data are in agreement with the idea that there is a size-dependency for the incorporation of D-amino acids and that an elongated linker to DNP can have positive effects on antibody recruitment.

Figure 5. UV−vis profile of the crude clarified peptidoglycan isolated from B. subtilis cells incubated with 500 μM of specified amino acid. 1484

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

Figure 6. (a) S. aureus were incubated for 4 h in LB alone or LB supplemented with 1 mM of D-Lys(DNP) followed by Alexa Fluor 488-conjugated anti-DNP antibodies in the presence of protein A and analyzed using flow cytometry. (b) S. aureus were incubated for 4 h in LB (containing 0.4 μg/ mL of tunicamycin) with either 1 mM of D-Lys(DNP) and/or protein A. Data are represented as mean + SD (n = 3).

Subsequently, we set out to show that the DART concept is compatible with a pathogenic Gram-positive bacterium. The ability of D-Lys(DNP)-remodeled surface of S. aureus to recruit anti-DNP antibody was evaluated (Figure 6). For this set of experiments, it was expected that endogenous S. aureus protein A would lead to IgG binding to the cell surface, thus contributing to background fluorescence. The addition of soluble protein A to the anti-DNP IgG antibody prior to incubation with the bacteria cells was successful in greatly reducing the non-DNP mediated IgG antibody binding. When S. aureus was labeled with 1 mM D-Lys(DNP), we observed a marked increase in anti-DNP antibody recruitment compared to unlabeled cells. Additionally, we hypothesized that teichoic acid−based polyanionic polymers (wall teichoic acid and lipoteichoic acid) contained within the cell wall of S. aureus may serve to hinder the permeation and binding of anti-DNP antibodies to the modified peptidoglycan.44,45 By growing the bacteria in the presence of tunicamycin, a known wall tecichoic acid inhibitor, antibody binding to the surface of the bacteria was greatly increased to levels far above the anti-DNP binding observed in previous experiments, thus demonstrating that our strategy has the potential to target pathogenic bacteria.46 We set out to confirm that D-Lys(DNP) was not toxic to host mammalian cells. One of the major possible advantages of using

D-Lys(DNP) to target bacteria comes from the fact that it relies on the enzymatic incorporation at the surface of the cell. No mammalian enzymes are known to exist that accept D-amino acid substrates, potentially giving this strategy high selectivity. Furthermore, the zwitterionic character of the free amino acid should greatly reduce its ability to passively diffuse into cells. Therefore, it is anticipated that in a living organism, administered D-amino acids would be preferentially utilized by bacteria. To evaluate the toxicity of D-Lys(DNP), mammalian cells (HeLa) were incubated in the presence of varying concentrations of D-Lys(DNP) for 72 h and analyzed by standard viability assays (Figure 7). At all concentrations examined, we observed no significant loss of viability, thus highlighting an important feasibility consideration of our strategy. Finally, we investigated the ability of anti-DNP opsonized bacterial cells to induce a specific response from immunocomponents. In the general view of antimicrobial activity of the innate immune system, the binding of antibodies to the surface of the bacteria can lead to the clearance of the pathogen by antibody-specific binding of macrophages via Fc receptors and subsequent opsonic phagocytosis, complement activation mediated by complement-dependent cytotoxicity (CDC), and antibody dependent cellular toxicity (ADCC).47 It is well

1485

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

Figure 7. Mammalian cell viability. HeLa cells were incubated for 72 h in the absence or the presence of increasing concentrations of DLys(DNP). Cellular viability was evaluated with MTT by measuring absorbance at 580 nm. Data are represented as mean + SD (n = 3).

Figure 8. Phagocytosis of bacterial cells. Phagocytosis of bacterial cells were evaluated following an overnight incubation period with 1 mM DLys(DNP) or in the absence of the D-amino acid. Untreated or opsonized cells (with anti-DNP antibody) were incubated with J774A.1 macrophages for 20 min in the absence or the presence of calcein-AM labeled B. subtilis cells. Fluorescence was measured by flow cytometry. Data are represented as mean + SD (n = 3).

established that Gram-positive bacteria are inherently resistant to the actions of the complement system. The presence of a thick layer of peptidoglycan on the cell wall surface is believed to inhibit the actions of the membrane attack complex (MAC).48,49 Likewise, we observed no killing of B. subtilis both unlabeled and labeled with 1 mM of D-Lys(DNP) upon the incubation of up to 80% pooled human serum (depleted of lysozyme by bentonite) during a 3 h incubation at 37 °C (data not shown). The degree of involvement of natural killer cell mediated cytotoxicity of opsonized Gram-positive bacteria is poorly understood. Although activated lymphocytes and natural killer cells can utilize cytotoxic proteins and peptides to induce an antimicrobial response in the presence of these pathogens, it is unclear to what extent this activity is coupled to the opsonization of Gram-positive bacteria.50 On the contrary, the recognition and phagocytosis of opsonized bacteria by macrophages has been widely recognized as the predominant mode of bacterial clearance upon primary antibody binding.25,29,51,52 The surface of B. subtilis was remodeled overnight with D-Lys(DNP) followed by the opsonization with anti-DNP antibodies. Additionally, B. subtilis were labeled with calceinAM to facilitate the tracking of the bacteria into macrophages. Incubation of opsonized B. subtilis in the presence of macrophages led to a 2-fold increase in phagocytosis compared to unlabeled B. subtilis (Figure 8). The uptake of unlabeled cells by macrophages is likely due to the size of the bacteria and their ability to trigger phagocytosis by engaging endogenous toll-like receptors. No increase in uptake was observed with the introduction of anti-DNP antibodies to unlabeled B. subtilis, an indication that there is little nonspecific binding of anti-DNP to the bacterial surface. We anticipate that the unique features of DART may reduce the potential for the emergence of resistance. The coupling of the antigen to a nonessential surface-bound protein could have left open the possibility of the bacteria rapidly evolving resistance. Instead, we utilized a surface modification method whereby the antigen is covalently conjugated to an essential three-dimensional scaffold. Furthermore, peptidoglycan remodeling should generate a surface with a high-valency of DNP epitopes, and, therefore, a robust recruitment of antibodies. Finally, the propensity of bacteria from across various phylogenetic groups to incorporate unnatural D-amino acids hints at a well conserved process. It is reasonable to expect that

this unusual property of transpeptidases is critical for the survival and proliferation of bacteria, and, therefore, less prone to resistance-conferring mutations. One outstanding question related to the D-amino acid mediated labeling of bacteria as it pertains to its therapeutic potential will be its specificity for pathogenic bacteria versus beneficial bacteria. To date, only pathogenic bacteria have been evaluated for their susceptibility for unnatural D-amino acid labeling. We anticipate that some beneficial bacteria may also be labeled with DNP-displaying Damino acids in the strategy described here. However, as with other molecules that indiscriminately target transpeptidasemediated enzymatic activities (e.g., penicillin and its many derivatives), it may be the case that lack of specificity may not prevent therapeutic activity. We plan to explore the peptidoglycan remodeling specificity in the future by using live organism models infected with pathogenic bacteria. A feature that renders bacterial infections difficult to treat is the presence of persister cells−cells that are in a quiescent-like state.53−55 Persister cells include slow- and nondividing cells that have altered metabolic states. Persister bacterial cells are less susceptible to the majority of traditional antibiotics as these agents target the processes involved in actively growing bacteria. We infer from the elevated levels of antibody recruitment by cells labeled in the stationary phase that slowand nondividing bacteria may, in fact, be more prone to the incorporation of D-Lys(DNP) than rapidly dividing bacteria (Figure S10, Supporting Information). Moreover, we observed detectable antibody recruitment at medium concentrations of D-Lys(DNP) down to 50 μM upon overnight incubation as opposed to the typical 4 h incubation used for all other labeling experiments (Figure S11, Supporting Information). The slower turnover in peptidoglycan from slow-dividing or persister cells combined with extended exposure to antibody recruiting Damino acids may result in an enhanced clearance potential. Moreover, a slow peptidoglycan turnover may allow for the use of reduced extracellular dosages of the D-amino acid required to attain a critical level of DNP displayed on the cell surface. 1486

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

from light and incubated at 37 °C for 1 h. Samples were then immediately analyzed by flow cytometry. Phagocytosis of Opsonized B. Subtilis. B. subtilis were inoculated (1:100) in LB broth supplemented with 1 mM DLys(DNP) and allowed to grow overnight at 37 °C with shaking. After washing with PBS, the cells were resuspended in PBS containing 10 μM calcein-AM and incubated at 37 °C for 30 min. The bacteria were then washed three times with PBS and incubated with 20 μg/mL of rabbit antidinitrophenyl IgG-fraction KLH (2 mg mL−1, Life Technologies, catalog no. A6430) and 10% (v/v) heat inactivated FBS in PBS to opsonize the bacteria. J774A.1 cells were cultured at 37 °C (in the presence of 5% CO2) in DMEM medium supplemented with 50 μg/mL streptomycin, 50 IU/mL penicillin, 2 mM L-glutamine, and 10% heat inactivated FBS. On the day of the experiments, J774 cells were washed twice with Balanced Salt Solution (BSS) by centrifuging 5 min at 250g at 4 °C. The washed J774A.1 cells were then mixed with opsonized B. subtilis in a ratio of 1:5 and incubated in BSS containing 5% FBS. The cell mixtures were then rotated at 37 °C for 20 min to induce phagocytosis, washed three times with cold BSS, and fixed for 30 min in PBS with 4% paraformaldehyde. Following fixation, the cells were resuspended in 0.5 mL PBS and analyzed by flow cytometry. For flow cytometry analysis, cells were analyzed using a BDFacs Canto II flow cytometer (BD Biosciences, San Jose, CA) equipped with a 488 nm argon laser and a 530/30 band-pass filter (FL1). The fluorescence data are expressed as mean arbitrary fluorescence units and were gated to include all J774A.1 cells. Peptidoglycan Isolation. 50 mL of B. subtilis bacteria were grown at 37 °C OD600 0.6 in LB broth, at which point the medium was replaced with LB broth supplemented with 500 μM of either DDab(DNP), D-Lys(DNP), or D-Lys-PEG1(DNP). The cells were allowed to incubate at 37 °C for 4 h in this medium before being harvested and washed with PBS (3×, 50 mL each). The cells were then resuspended in PBS and boiled for 7 min and then centrifuged at 14 000g for 8 min at 4 °C. Cells were then placed in 25 mL of 5% (w/v) sodium dodecyl sulfate (SDS) and boiled for 25 min followed by centrifugation at 14,000g for 8 min at 4 °C. Following centrifugation, cells were boiled again in 25 mL of 4% (w/v) SDS for 15 min followed by centrifugation using same parameters as before. Cells were then washed 5 times with 60 °C DI water to remove all SDS. After washing, cells were incubated in 6 mL of 50 mM Tris HCl and 2 mg mL−1 Proteinase K for 1 h at 60 °C, and then washed 3 times with DI water. The cell wall pellet was then resuspended and digested with 250 μg/ mL lysozyme in 25 mM sodium phosphate buffer pH 5.6 for 15 h at 37 °C. The digestion was then ceased by boiling for 3 min. Incorporation was analyzed using a Shimazdo UV-2101PC, PerkinElmer Series 200 HPLC, and Bruker Microflex MALDI-TOF MS. MTT Cell Viability Assay. Cells were cultured at 37 °C with 5% CO2. HeLa cells were cultured in DMEM medium supplemented with 10% fetal bovine serum (Cambrex Bio Science Walkersville, Inc.), 2 mM L-glutamine (Cellgro, Mediatech), and 50 units/mL penicillin and 50 μg/mL streptomycin (Cellgro, Mediatech). HeLa cells were seeded at a density of 3000 cells per well in a 96 well plate and incubated overnight at 37 °C with 5% CO2 in DMEM medium. After 24 h, the medium was replaced with DMEM medium supplemented with varying concentrations of D-Lys(DNP) and incubated for 72 h. After incubation, the medium was removed and cells were treated with 100 μL of 0.5 mg mL−1 3-(4,5-dimethylthiazol-2yl)-2,5 diphenyltetrazolium bromide (MTT) in the dark for 2 h. The MTT solution was then removed and the formazan crystals were resolubilized in 200 μL DMSO. The absorbance was read on a plate reader at 580 nm to determine cell viability. Bacteria Growth Curve. Overnight cultures of B. subtilis were diluted 1:100 in LB growth medium in the presence or absence of the appropriate concentration of D-Lys(DNP) at 37 °C. Bacteria were grown in culture flasks and OD600 was recorded every 30 min to determine bacterial viability in the presence of D-Lys(DNP).

Conclusion. In conclusion, we demonstrated the initial steps toward the development of DART, a novel synthetic immunology approach for remodeling the surface of Grampositive bacteria for clearance by the host immune system. This strategy utilizes the inherent promiscuity of bacteria to incorporate extracellular D-amino acids into their peptidoglycan. We show that the strategy is successful in specifically recruiting antibodies to the surface of B. subtilis and S. aureus and may be a unique platform for the development of nextgeneration antimicrobial therapies.

■

METHODS

Materials. Reagents and Bacterial strains. Amino acids were purchased from Chem-Impex. Antibody reagents were purchased from Life Technologies. Protein A was purchased from Sigma-Aldrich. All other organic reagents were purchased from Fisher Scientific and used without further purification. B. subtilis ΔdacA and S. aureus ATCC 25923 were the strains of bacteria used for the performed experiments. B. subtilis D-Amino Fluorescent Labeling. B. subtilis was grown at 37 °C to an OD600 0.6 in LB broth, at which point the mediamedium was replaced with LB broth supplemented with either D- or L-Lys(NBD). The bacteria were then incubated at 37 °C overnight or for the desired time, harvested, washed three times with PBS, fixated with 2% formaldehyde solution and analyzed via flow cytometry on a BDFacs Canto II or imaged with Nikon Eclipse TE2000-U microscope at 100× magnification. For inhibition of DLys(NBD) incorporation, the medium was additionally supplemented with varying concentrations of D-Ala when incubating bacteria in the presence D-Lys(NBD). B. subtilis D-Amino Acid Antibody Binding Assay. B. subtilis bacteria were grown at 37 °C to OD600 0.6 in LB broth, at which point the medium was replaced with LB broth containing 250 μM of the appropriate D-amino acid. The bacteria were allowed to incubate for 4 h at 37 °C before being harvested and washed three times with phosphate buffer saline (PBS). Approximately 2 × 106 cfu were then incubated in 100 μL of PBS containing 10% (v/v) FBS and 1 μL of Alexa Fluor 488 conjugated rabbit antidinitrophenyl IgG-fraction KLH (2 mg mL−1, Life Technologies, catalog no. A11097). All experiments were protected from light and incubated at 37 °C for 1 h. Samples were then immediately analyzed by flow cytometry or imaged using fluorescence microscopy. For flow cytometry analysis, cells were analyzed using a BDFacs Canto II flow cytometer (BD Biosciences, San Jose, CA) equipped with a 488 nm argon laser and a 530 bandpass filter (FL1). A minimum of 10 000 events were counted for each data point. The data was analyzed using the FACSDiva version 6.1.1 software. The fluorescence data are expressed as mean arbitrary fluorescence units and were gated to include all healthy bacteria. For exponential phase labeling: Bacteria were grown at 37 °C to an OD600 0.6 before harvesting and replacing the medium with D-amino acid supplemented LB. For stationary phase labeling: Bacteria were grown at 37 °C to an OD600 1.5 before harvesting and replacing the medium with D-amino acid supplemented LB. For bacteria labeled with DNPdisplaying amino acids, the background fluorescence (unlabeled bacteria incubated with anti-DNP) was subtracted from the final values. S. aureus D-Amino Acid Antibody Binding Assay. S. aureus bacteria were grown overnight at 37 °C in LB broth either in the presence or absence of 0.4 μg/mL tunicamycin. The medium was then replaced with LB broth containing 1 mM D-Lys(DNP) and the bacteria were allowed to incubate at 37 °C for 4 h before being harvested and washed three times with phosphate buffer saline (PBS). Approximately 2 × 106 cfu were then incubated in 100 μL of PBS containing 10% (v/v) FBS and 1 μL of Alexa Fluor 488 conjugated rabbit antidinitrophenyl IgG-fraction KLH (2 mg mL−1, Life Technologies, catalog no. A11097). To decrease background fluorescence from native S. aureus protein A binding the IgG antibody, these solutions were supplemented with soluble S. aureus protein A to a final concentration of 1.6 mg mL−1. All experiments were protected 1487

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

■

Articles

vancomycin-resistant bacterial cell-wall precursors: UDP-N-acetylmuramyl-peptides. J. Am. Chem. Soc. 123, 9916−9917. (16) Sadamoto, R., Niikura, K., Ueda, T., Monde, K., Fukuhara, N., and Nishimura, S. (2004) Control of bacteria adhesion by cell-wall engineering. J. Am. Chem. Soc. 126, 3755−3761. (17) Kuru, E., Hughes, H. V., Brown, P. J., Hall, E., Tekkam, S., Cava, F., de Pedro, M. A., Brun, Y. V., and VanNieuwenhze, M. S. (2012) In situ probing of newly synthesized peptidoglycan in live bacteria with fluorescent D-amino acids. Angew. Chem., Int. Ed. Engl. 51, 12519− 12523. (18) Liechti, G. W., Kuru, E., Hall, E., Kalinda, A., Brun, Y. V., VanNieuwenhze, M., and Maurelli, A. T. (2014) A new metabolic cellwall labelling method reveals peptidoglycan in Chlamydia trachomatis. Nature 506, 507−510. (19) Pilhofer, M., Aistleitner, K., Biboy, J., Gray, J., Kuru, E., Hall, E., Brun, Y. V., VanNieuwenhze, M. S., Vollmer, W., Horn, M., and Jensen, G. J. (2013) Discovery of chlamydial peptidoglycan reveals bacteria with murein sacculi but without FtsZ. Nat. Commun. 4, 2856. (20) Siegrist, M. S., Whiteside, S., Jewett, J. C., Aditham, A., Cava, F., and Bertozzi, C. R. (2013) D-Amino acid chemical reporters reveal peptidoglycan dynamics of an intracellular pathogen. ACS Chem. Biol. 8, 500−505. (21) Lam, H., Oh, D. C., Cava, F., Takacs, C. N., Clardy, J., de Pedro, M. A., and Waldor, M. K. (2009) D-Amino acids govern stationary phase cell wall remodeling in bacteria. Science 325, 1552−1555. (22) Nelson, J. W., Chamessian, A. G., McEnaney, P. J., Murelli, R. P., Kazmierczak, B. I., and Spiegel, D. A. (2010) A biosynthetic strategy for re-engineering the Staphylococcus aureus cell wall with non-native small molecules. ACS Chem. Biol. 5, 1147−1155. (23) Lupoli, T. J., Tsukamoto, H., Doud, E. H., Wang, T. S., Walker, S., and Kahne, D. (2011) Transpeptidase-mediated incorporation of Damino acids into bacterial peptidoglycan. J. Am. Chem. Soc. 133, 10748−10751. (24) Carlson, C. B., Mowery, P., Owen, R. M., Dykhuizen, E. C., and Kiessling, L. L. (2007) Selective tumor cell targeting using low-affinity, multivalent interactions. ACS Chem. Biol. 2, 119−127. (25) Krishnamurthy, V. M., Quinton, L. J., Estroff, L. A., Metallo, S. J., Isaacs, J. M., Mizgerd, J. P., and Whitesides, G. M. (2006) Promotion of opsonization by antibodies and phagocytosis of Grampositive bacteria by a bifunctional polyacrylamide. Biomaterials 27, 3663−3674. (26) Lu, Y., You, F., Vlahov, I., Westrick, E., Fan, M., Low, P. S., and Leamon, C. P. (2007) Folate-targeted dinitrophenyl hapten immunotherapy: Effect of linker chemistry on antitumor activity and allergic potential. Mol. Pharmaceutics 4, 695−706. (27) Owen, R. M., Carlson, C. B., Xu, J., Mowery, P., Fasella, E., and Kiessling, L. L. (2007) Bifunctional ligands that target cells displaying the α vs β3 integrin. Chembiochem 8, 68−82. (28) Rader, C., Sinha, S. C., Popkov, M., Lerner, R. A., and Barbas, C. F., 3rd (2003) Chemically programmed monoclonal antibodies for cancer therapy: adaptor immunotherapy based on a covalent antibody catalyst. Proc. Natl. Acad. Sci. U.S.A. 100, 5396−5400. (29) Bertozzi, C. R., and Bednarski, M. D. (1992) A ReceptorMediated Immune-Response Using Synthetic Glycoconjugates. J. Am. Chem. Soc. 114, 5543−5546. (30) Li, J., Zacharek, S., Chen, X., Wang, J. Q., Zhang, W., Janczuk, A., and Wang, P. G. (1999) Bacteria targeted by human natural antibodies using α-Gal conjugated receptor-specific glycopolymers. Bioorg. Med. Chem. 7, 1549−1558. (31) Farah, F. S. (1973) Natural antibodies specific to the 2,4dinitrophenyl group. Immunology 25, 217−226. (32) Murelli, R. P., Zhang, A. X., Michel, J., Jorgensen, W. L., and Spiegel, D. A. (2009) Chemical control over immune recognition: A class of antibody-recruiting small molecules that target prostate cancer. J. Am. Chem. Soc. 131, 17090−17092. (33) Parker, C. G., Domaoal, R. A., Anderson, K. S., and Spiegel, D. A. (2009) An antibody-recruiting small molecule that targets HIV gp120. J. Am. Chem. Soc. 131, 16392−16394.

ASSOCIATED CONTENT

S Supporting Information *

Additional experimental details (synthesis of small molecules) and figures. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Email: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was supported by Lehigh University (M.P.). We thank D. Popham (Virginia Institute of Technology) and D. Kearns (Indiana University) for the kind gifts of B. subtilis and B. subtilis ΔdacA strains and J. Chmielewski (Purdue University) for the kind gift of S. aureus.

■

REFERENCES

(1) Fisher, J. F., Meroueh, S. O., and Mobashery, S. (2005) Bacterial resistance to β-lactam antibiotics: Compelling opportunism, compelling opportunity. Chem. Rev. 105, 395−424. (2) Levy, S. B., and Marshall, B. (2004) Antibacterial resistance worldwide: Causes, challenges, and responses. Nat. Med. 10, S122− 129. (3) Lewis, K. (2007) Persister cells, dormancy and infectious disease. Nat. Rev. Microbiol. 5, 48−56. (4) Ambrose, P. G., Hammel, J. P., Bhavnani, S. M., Rubino, C. M., Ellis-Grosse, E. J., and Drusano, G. L. (2012) Frequentist and Bayesian pharmacometric-based approaches to facilitate critically needed new antibiotic development: Overcoming lies, damn lies, and statistics. Antimicrob. Agents Chemother. 56, 1466−1470. (5) Cegelski, L., Marshall, G. R., Eldridge, G. R., and Hultgren, S. J. (2008) The biology and future prospects of antivirulence therapies. Nat. Rev. Microbiol. 6, 17−27. (6) Nathan, C. (2012) Fresh approaches to anti-infective therapies. Sci. Transl. Med. 4, 140−142. (7) Oka, T., Hashizume, K., and Fujita, H. (1980) Inhibition of peptidoglycan transpeptidase by β-lactam antibiotics: Structure− activity relationships. J. Antibiot. 33, 1357−1362. (8) van Heijenoort, J. (2001) Recent advances in the formation of the bacterial peptidoglycan monomer unit. Nat. Prod. Rep. 18, 503− 519. (9) Kohanski, M. A., Dwyer, D. J., and Collins, J. J. (2010) How antibiotics kill bacteria: From targets to networks. Nat. Rev. Microbiol. 8, 423−435. (10) Gautam, S., Gniadek, T. J., Kim, T., and Spiegel, D. A. (2013) Exterior design: Strategies for redecorating the bacterial surface with small molecules. Trends Biotechnol. 31, 258−267. (11) Cohen, A. S., Dubikovskaya, E. A., Rush, J. S., and Bertozzi, C. R. (2010) Real-time bioluminescence imaging of glycans on live cells. J. Am. Chem. Soc. 132, 8563−8565. (12) Kaewsapsak, P., Esonu, O., and Dube, D. H. (2013) Recruiting the host’s immune system to target Helicobacter pylori’s surface glycans. ChemBioChem 14, 721−726. (13) Sadamoto, R., Matsubayashi, T., Shimizu, M., Ueda, T., Koshida, S., Koda, T., and Nishimura, S. (2008) Bacterial surface engineering utilizing glucosamine phosphate derivatives as cell wall precursor surrogates. Chem.Eur. J. 14, 10192−10195. (14) Sadamoto, R., Niikura, K., Sears, P. S., Liu, H., Wong, C. H., Suksomcheep, A., Tomita, F., Monde, K., and Nishimura, S. (2002) Cell-wall engineering of living bacteria. J. Am. Chem. Soc. 124, 9018− 9019. (15) Liu, H., Sadamoto, R., Sears, P. S., and Wong, C. H. (2001) An efficient chemoenzymatic strategy for the synthesis of wild-type and 1488

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489

ACS Chemical Biology

Articles

(34) Karjalainen, K., and Makela, O. (1976) Concentrations of three hapten-binding immunoglobulins in pooled normal human serum. Eur. J. Immunol. 6, 88−93. (35) Jakobsche, C. E., Parker, C. G., Tao, R. N., Kolesnikova, M. D., Douglass, E. F., Jr., and Spiegel, D. A. (2013) Exploring binding and effector functions of natural human antibodies using synthetic immunomodulators. ACS Chem. Biol. 8, 2404−2411. (36) Sonenshein, A. L., Hoch, J. A., and Losick, R. (1993) Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology, and Molecular Genetics; American Society for Microbiology, Washington, DC. (37) Popham, D. L., Gilmore, M. E., and Setlow, P. (1999) Roles of low-molecular-weight penicillin-binding proteins in Bacillus subtilis spore peptidoglycan synthesis and spore properties. J. Bacteriol. 181, 126−132. (38) Buchanan, C. E., and Gustafson, A. (1992) Mutagenesis and mapping of the gene for a sporulation-specific penicillin-binding protein in Bacillus subtilis. J. Bacteriol. 174, 5430−5435. (39) Kolodkin-Gal, I., Romero, D., Cao, S., Clardy, J., Kolter, R., and Losick, R. (2010) D-Amino acids trigger biofilm disassembly. Science 328, 627−629. (40) Vollmer, W., Blanot, D., and de Pedro, M. A. (2008) Peptidoglycan structure and architecture. FEMS Microbiol. Rev. 32, 149−167. (41) Meroueh, S. O., Bencze, K. Z., Hesek, D., Lee, M., Fisher, J. F., Stemmler, T. L., and Mobashery, S. (2006) Three-dimensional structure of the bacterial cell wall peptidoglycan. Proc. Natl. Acad. Sci. U.S.A. 103, 4404−4409. (42) Vollmer, W., Blanot, D., and de Pedro, M. A. (2008) Peptidoglycan structure and architecture. FEMS Microbiol. Rev. 32, 149−167. (43) Atrih, A., Bacher, G., Allmaier, G., Williamson, M. P., and Foster, S. J. (1999) Analysis of peptidoglycan structure from vegetative cells of Bacillus subtilis 168 and role of PBP 5 in peptidoglycan maturation. J. Bacteriol. 181, 3956−3966. (44) Brown, S., Santa Maria, J. P., Jr., and Walker, S. (2013) Wall teichoic acids of gram-positive bacteria. Annu. Rev. Microbiol. 67, 313− 336. (45) Swoboda, J. G., Campbell, J., Meredith, T. C., and Walker, S. (2010) Wall teichoic acid function, biosynthesis, and inhibition. Chembiochem 11, 35−45. (46) Campbell, J., Singh, A. K., Santa Maria, J. P., Jr., Kim, Y., Brown, S., Swoboda, J. G., Mylonakis, E., Wilkinson, B. J., and Walker, S. (2011) Synthetic lethal compound combinations reveal a fundamental connection between wall teichoic acid and peptidoglycan biosyntheses in Staphylococcus aureus. ACS Chem. Biol. 6, 106−116. (47) Burton, D. R. (1990) Antibody: The flexible adaptor molecule. Trends Biochem. Sci. 15, 64−69. (48) Joiner, K., Brown, E., Hammer, C., Warren, K., and Frank, M. (1983) Studies on the mechanism of bacterial resistance to complement-mediated killing. III. C5b-9 deposits stably on rough and type 7 S. pneumoniae without causing bacterial killing. J. Immunol. 130, 845−849. (49) Joiner, K. A. (1988) Complement evasion by bacteria and parasites. Annu. Rev. Microbiol. 42, 201−230. (50) Tarkkanen, J., and Saksela, E. (1991) Potentiation of human natural-killer-cell cytotoxicity by salmonella bacteria is an interferon-2independent and interleukin-2-independent process that utilizes CD2 and CD18 structures in the effector phase. Infect. Immun. 59, 2767− 2773. (51) Aderem, A., and Underhill, D. M. (1999) Mechanisms of phagocytosis in macrophages. Annu. Rev. Immunol. 17, 593−623. (52) Segura, M., and Gottschalk, M. (2002) Streptococcus suis interactions with the murine macrophage cell line J774: Adhesion and cytotoxicity. Infect. Immun. 70, 4312−4322. (53) Hurdle, J. G., O’Neill, A. J., Chopra, I., and Lee, R. E. (2011) Targeting bacterial membrane function: An underexploited mechanism for treating persistent infections. Nat. Rev. Microbiol. 9, 62−75.

(54) Levin, B. R., and Rozen, D. E. (2006) Non-inherited antibiotic resistance. Nat. Rev. Microbiol. 4, 556−562. (55) Lewis, K. (2010) Persister cells. Annu. Rev. Microbiol. 64, 357− 372.

1489

dx.doi.org/10.1021/cb5002685 | ACS Chem. Biol. 2014, 9, 1480−1489